Metastatic pancreatic cancer is a lethal disease. While most of the critical genetic and epigenetic alterations have been known for years, to date this has not resulted in useful therapeutics. Our most important goal is to identify alterations in PDAC gene expression that can be utilized to develop effective therapies. To this end we have recently been focusing on the transcription factor FoxM1, which is drastically increased in PDAC. To understand the importance of this elevated expression we have developed PDAC cell lines genetically engineered to over-express FoxM1. These cell lines uniformly exhibit greatly elevated growth and metastasis. In a complimentarily approach, we have reduced FoxM1 expression in numerous PDAC cells and observed a reduction in the aggressive nature of these cells. Recently, we discovered that human PDAC cells almost exclusively express a splice form of FoxM1 called FoxM1C, which is not observed in normal tissues. Based on our preliminary studies, we postulate that during PDAC carcinogenesis, dysregulated miRNA expression leads to overexpression of FoxM1 and dysregulated expression of its downstream target genes key to invasion and metastasis, resulting in an enhanced malignant potential of PDAC cells and cause poor prognosis of PDAC patients. Thus, FoxM1 is a malignant biomarker and therapeutic target in PDAC. To test our hypothesis, we propose the following three specific aims. Aim #1. Investigate the molecular mechanisms underlying the dysregulated FoxM1 expression in pancreatic cancer. The overexpression of FoxM1 in PDAC is well established, while the underlying mechanisms are unknown. We have recently utilized a computer based algorithm to screen for possible microRNAs that target FoxM1 (both in silico analysis and microarray analysis of miRNA expression in normal pancreas vs. pancreatic cancer) and have tentatively identified 3 FoxM1-related microRNAs. Interestingly, all three are down-regulated in pancreatic cancer. We expect that those microRNAs causally regulate FoxM1 expression and function and exhibit prognostic values. Aim #2. Determine the regulatory role of FoxM1 expression in pancreatic cancer invasion and metastasis. Human pancreatic cancer cells are known to overexpress uPAR and its expression levels directly correlate with those of FoxM1. Altered expression of uPAR regulates invasion and metastasis of pancreatic cancer. Therefore, we wish to determine whether uPAR is a novel important downstream target and functional mediator of FoxM1. We expect that FoxM1 isoforms regulate the expression and function of uPAR in pancreatic cancer cells and this novel pathway is essential to pancreatic cancer invasion and metastasis. Aim #3. Determine the clinical significance of FoxM1 isoforms in human pancreatic cancer invasion and metastasis. FoxM1 consists of 3 isoforms: FoxM1A, FoxM1B, and FoxM1C, and most recently, two additional isoforms are identified and provisionally named as FoxM1B1 and FoxM1B2. The expression and functions of those isoforms in pancreatic cancer are not completely known. Our preliminary studies have indicated that pancreatic cancer cells predominantly express FoxM1C and that this form has unique properties. We expect that the FoxM1C form is associated with particularly aggressive disease and altered expression of different isoforms of FoxM1, including FoxM1B1 and FoxM1B2 to FoxM1B, directly impact on pancreatic cancer growth and metastasis.